luma.gl

TransformFeedback (WebGL2)

TransformFeedback objects holds state needed to perform transform feedback operations. They store the buffer bindings that are being recorded to. This makes it easy to switch between different sets of feedback buffer bindings (somewhat similar to how VertexArrayObjects hold input vertex buffers.

The state managed by TransformFeedback objects includes the buffers the GPU will use to record the requested varyings.

When TransformFeedback objects must be “activated” (TransformFeedback.begin) before it can be used. There a number of caveats to be aware of when manually managing TransformFeedback object activation, see the remarks. For this reason, luma.gl Program.draw call takes an optional TransformFeedback object as a parameter and activates and deactivates it before and after the draw call.

Finally, note that when using transform feedback it is frequently desirable to turn off rasterization: gl.enable(GL.RASTERIZER_DISCARD) to prevent the fragment shader from running.

For more information, see OpenGL Wiki.

Usage

Setting up a model object for transform feedback.

const model = new Model(gl, {
  vs,
  fs,
  varyings: ['gl_Position', 'outputColor'],
  ...
});

Setting up a transform feedback object and binding buffers

const transformFeedback = new TransformFeedback(gl)
  .setBuffer(0, bufferPosition)
  .setBuffer(1, bufferColor);

When binding the buffers, index should be equal to the corresponding varying entry in varyings array passed to Program constructor.

Buffers can also be bound using varying name if information about varyings are retrieved from Program object.

const transformFeedback = new TransformFeedback(gl, {
  program: ..., // linked program, configuration will be read from it
  buffers: {
    outputColor: bufferColor,
    gl_Position: bufferPosition
  }
});

Running program (drawing) with implicit activation of transform feedback (will call begin and end on supplied transformFeedback)

model.draw({
  drawMode,
  vertexCount,
  ...,
  transformFeedback
});

Running a transform feedback operation while turning off rasterization (drawing):

model.transform({
  drawMode,
  ...,
  transformFeedback
});

or equivalently, just call draw with an additional parameter:

const parameters = {[GL.RASTERIZER_DISCARD]: true}
model.draw({..., transformFeedback, parameters});

Methods

constructor(gl : WebGL2RenderingContext, props: Object)

See TransformFeedback.setProps for parameters.

WebGL APIs gl.createTransformFeedback

initialize(props : Object) : TransformFeedback

Reinitializes an existing TransformFeedback object with new props.

setProps(props : Object) : TransformFeedback

• props.program= (Object) - Gets a mapping of varying name to buffer indices from a linked program if supplied.
• props.buffers=(Object) - Map of location index or name to Buffer object or buffer parameters object. If buffer parameters object is supplied, it contains following fields.
  • buffer=(Buffer) - Buffer object to be bound.
  • byteOffset=(Number, default: 0) - Byte offset that is used to start recording the data in the buffer.
  • byteSize=(Number, default: remaining buffer size) - Size in bytes that is used for recording the data.
• props.bindOnUse=true - If true, binds and unbinds buffers before and after use, rather than right away when set. Workaround for a possible Khronos/Chrome bug.

Notes:

• buffers - will get bound to indices in the GL.TRANSFORM_FEEDBACK_BUFFER target.

delete() : TransformFeedback

Destroys a TransformFeedback object.

WebGL APIS gl.deleteTransformFeedback

setBuffers(buffers: Object) : TransformFeedback

• buffers=(Object) - Map of location index or name to Buffer object or buffer parameters object. If buffer parameters object is supplied, it contains following fields.
  • buffer=(Buffer) - Buffer object to be bound.
  • byteOffset=(Number, default: 0) - Byte offset that is used to start recording the data in the buffer.
  • byteSize=(Number, default: remaining buffer size) - Size in bytes that is used for recording the data.

Notes:

• To use gl.bindBufferRange, either offsetInByts or byteSize must be specified, when only one is specified, default value is used for the other, when both not specified, gl.bindBufferBase is used for binding.

WebGL APIs gl.bindBufferBase, gl.bindBufferRange

begin(primitiveMode : GLEnum) : TransformFeedback

Activates transform feedback using the buffer bindings in this TransformFeedback object.

• primitiveMode (GLenum) -

returns (TransformFeedback) - returns self to enable chaining

Notes:

• Buffers can not be accessed until TransformFeedback.end or TransformFeedback.pause have been called.
• Buffers can not be changed until TransformFeedback.end or has been called, which includes doing anything which reads from or writes to any part of these buffers (outside of feedback writes, of course, or reallocating storage for any of these buffers).

WebGL APIs gl.beginTransformFeedback

end() : TransformFeedback

returns (TransformFeedback) - returns self to enable chaining

WebGL APIs gl.endTransformFeedback

See also

• Program constructor - varyings argument to specify which vertex shader outputs to expose to transform feedback operations.

Enumerations

Primitive Mode	Compatible Draw Modes
GL.POINTS	GL.POINTS
GL.LINES	GL.LINES, GL.LINE_LOOP, GL.LINE_STRIP
GL.TRIANGLES	GL.TRIANGLES, GL.TRIANGLE_STRIP, GL.TRIANGLE_FAN

Limits

Limit	Value	Description
GL.MAX_TRANSFORM_FEEDBACK_SEPARATE_ATTRIBS	>=4	total number of variables that can be captured }
GL.MAX_TRANSFORM_FEEDBACK_SEPARATE_COMPONENTS	>=4	number of components that any particular variable can contain
GL.MAX_TRANSFORM_FEEDBACK_INTERLEAVED_COMPONENTS	>= 64	total number of components in interleaved capture
GL.MAX_TRANSFORM_FEEDBACK_BUFFERS	TBD	Advanced interleaving total number of buffers

Remarks

About TransformFeedback activation caveats

• When activated, TransformFeedback are coupled to the “current” Program
• Note that a started and unpaused TransformFeedback prevents the app from changing or re-linking the current program. So for instance, Program.use (gl.useProgram) cannot be called.

This site is open source. Improve this page.